| In a sense, the rapid development of unmanned ground vehicles benefits from thepre-established test and evaluation system. The test and evaluation system has played animportant role and formulated the related technical indicators to guide the development ofunmanned ground vehicles.Depended on "Research on Comprehensive Test Environment Design and EvaluationArchitecture of Intelligent Behavior for Unmanned Ground Vehicles"(90920304) and"Design and Implementation of the Verification Environment for Cognitive Ability Tests ofUnmanned Ground Vehicles"(91120010) supported by National Natural ScienceFoundation of China, the research of the test and evaluation system has been carried out bystages, with the purpose of proposing a quantitative evaluation method for the intelligentlevel of unmanned ground vehicles. The evaluation index system was divided into fourevaluation levels established based on the complexity of the environment perception andthe intelligent behavioral decision-making of unmanned ground vehicles. The scientific andrational weights of evaluation indexes were determined by the extended AHP method. Thechaos theory was used for quantitative analysis of the evaluation index of unmannedground vehicles, so as to achieve the quantitative evaluation of the intelligent level ofunmanned ground vehicles.Firstly, the quantitative evaluation of the intelligent level of unmanned ground vehicleswas studied. The complex interactive system which was consisted of unmanned groundvehicles, tasks and environment was the object of the study. The evaluation model ofunmanned ground vehicles was established by the complexity of the task, the complexity ofthe environment and the degree of human intervention. The intelligent level of unmannedground vehicles was classified according to the complexity of the task, the complexity ofthe environment, the degree of human intervention and the quality of the running quality.Secondly, the relatively complete evaluation index system was established for theintelligent level of unmanned ground vehicles. Four evaluation levels: evaluation objectives,evaluation aspects, evaluation elements, evaluation factors, were divided based on thecomplexity of the environment perception and the intelligent behavioral decision-making ofunmanned ground vehicles. The multi-level evaluation index system was established withthe intelligent level of unmanned ground vehicles as the evaluation objective. Then, the quintic polynomial of the ideal trajectory planning method was proposedbased on the excellent driver’s trajectory. Considering traffic safety, the safe spacing of lanechange avoidance was deduced. The excellent driver’s trajectory was collected through themethod of GPS/DR integrated positioning, Then the initial states and the target states of thevehicle were obtained (considering the safe spacing). The ideal trajectory of unmannedground vehicles was planned by the five polynomial of the planning method.Furthermore, the weights of evaluation indexes were determined by the extended AHPmethod. The method overcomed the ambiguity problem of the AHP expertise judgment andthe consistency problem of the judgment matrix. Using extension theory with the intervalnumber instead of point values, the extension judgment matrix was constructed, whichmade the weights of evaluation indexes more reasonable.Afterward, the trajectory of unmanned ground vehicles was quantitative analyzedbased on chaos theory. The time delay and the embedding dimension of the deviation timeseries of unmanned ground vehicles were determined by the C-C method, then the phasespace of the deviation time series was reconstructed. The Lyapunov exponent of thedeviation time series was calculated by the wolf method. The quantitative representation ofthe trajectory of unmanned ground vehicles can realize quantitative evaluating of unmannedground vehicles, so as to lay a good foundation for improvement of the evaluation systemand establishment of the evaluation criteria.Finally, the chaos extension analytic hierarchy process (Chaos-EAHP) method wassuccessfully applied to quantitatively evaluate the intelligent level of unmanned groundvehicles. The Chaos-EAHP method can quantitatively evaluate the individual module andthe overall technical performance of unmanned ground vehicles, and it guides theunmanned ground vehicles not only to complete the task, but also to complete the task withhigh standard and high quality. |
